CtrlK
BlogDocsLog inGet started
Tessl Logo

prompt-review

This skill should be used when the user asks to "review the prompt", "audit the system prompt", "check prompt quality", "inspect what the LLM sees", "debug prompt issues", or "find prompt engineering problems". Pulls the live rendered prompt via the API, explains how it's composed, and reviews it for issues.

68

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

71%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A thorough, highly actionable audit skill with concrete commands, paths, and checklists. Its weakness is monolithic structure: the long reference material (directory tree, layer breakdown, rendering order) is inlined rather than progressively disclosed into separate reference files.

Suggestions

Move the full annotated directory layout and the 18-step rendering-order list into a references/ file (e.g. PROMPT_LAYERS.md), keeping SKILL.md as a concise overview that links to it — this improves both conciseness and progressive disclosure.

Add an inline validation checkpoint in Step 2 (e.g., confirm the inspect response contains a non-empty system_prompt before writing it to /tmp/prompt_inspect.md and reading it), since the Notes already flag that inactive channels are not inspectable.

Trim the per-layer 'Review focus' lines or fold them into the checklist to reduce length without losing the diagnostic guidance.

DimensionReasoningScore

Conciseness

Mostly efficient and assumes Claude's intelligence (no generic "what is a system prompt" padding), but at ~290 lines the body inlines a full annotated directory tree, an 18-step rendering order, and exhaustive per-layer descriptions that could be tightened or offloaded.

3 / 5

Actionability

Fully executable guidance: copy-paste curl+jq commands with concrete endpoints, specific file paths to read, a structured review checklist, and a problem-to-fix-location table mapping each finding to the exact file to edit.

5 / 5

Workflow Clarity

Clear five-step procedure (list channels, pull prompt, read files, review, report) with a detailed review checklist and structured report format; minor validation gaps — e.g., no inline check that the inspect endpoint returned a prompt before analyzing.

4 / 5

Progressive Disclosure

Well-organized with clear headers and a navigable structure, but no bundle files exist and all content — including the full directory layout and layer reference — is inlined in one ~290-line file rather than split into one-level-deep reference files.

3 / 5

Total

15

/

20

Passed

Description

95%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that clearly states both capability and trigger conditions with natural user phrasing. The only soft spot is that the trailing "reviews it for issues" action is slightly generic compared to the rest.

DimensionReasoningScore

Specificity

Names the domain and several concrete actions — "Pulls the live rendered prompt via the API, explains how it's composed, and reviews it for issues" — though "reviews it for issues" is somewhat generic, leaving minor coverage gaps versus a fully comprehensive action list.

4 / 5

Completeness

Explicitly answers both what ("Pulls the live rendered prompt... explains how it's composed, and reviews it for issues") and when ("This skill should be used when the user asks to...") with concrete trigger phrases.

5 / 5

Trigger Term Quality

Provides comprehensive natural-language triggers users would actually say — "review the prompt", "audit the system prompt", "check prompt quality", "inspect what the LLM sees", "debug prompt issues", "find prompt engineering problems" — covering synonyms and paraphrases well.

5 / 5

Distinctiveness Conflict Risk

Targets a clear niche — auditing a Spacebot agent's live rendered system prompt — with distinct triggers and minimal overlap risk with other skills.

5 / 5

Total

19

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
spacedriveapp/spacebot
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.